SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 03030 · ECCV · 2024

Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation

Ziyu Wang, Yong-Lu Li, Cewu Lu, Yu-Wing Tai, Chi-Keung Tang, Yue Xu, Kaitong Cui

Open in the Atlas

Code that ran

6 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 3 failures of our instrument; 2 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/03030.json · badge: /api/ran/03030.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
silicx/GoldFromOres-BiLP canonical (classifier label)
classifier basis: census
6 ran · 2 no recorded run
FunctionStatusWhere it lives
VGG11 Ran
ran; our contract draft was wrong, not the code
silicx/GoldFromOres-BiLP/DatasetCondensation/networks.py
pointer only (licence: MIT); the stored copy get_code serves, from ndhg1213/CMIDD, is pointer only (licence: NONE) · get_code("8e1b179ef1f0fe03")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ndhg1213/CMIDD, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

VGG11BN Ran
ran; our contract draft was wrong, not the code
silicx/GoldFromOres-BiLP/DatasetCondensation/networks.py
pointer only (licence: MIT); the stored copy get_code serves, from ndhg1213/CMIDD, is pointer only (licence: NONE) · get_code("cfab0fafbf988aa4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ndhg1213/CMIDD, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

VGG13 Ran
ran; our contract draft was wrong, not the code
silicx/GoldFromOres-BiLP/DatasetCondensation/networks.py
pointer only (licence: MIT); the stored copy get_code serves, from ndhg1213/CMIDD, is pointer only (licence: NONE) · get_code("07962bf28fe347a0")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ndhg1213/CMIDD, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

distance_wb Ran
ran, honoured the contract we drafted
silicx/GoldFromOres-BiLP/DatasetCondensation/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from saehyung-lee/dcc, is pointer only (licence: NONE) · get_code("6e9c65aaa7c1b5da")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from saehyung-lee/dcc, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

drop_samples Ran
ran on a synthesized input
silicx/GoldFromOres-BiLP/drop_utils/drop.py
code served (permissive licence) · get_code("55733ece87768334")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)
np.random.seed(0)

# Create tiny tensors for images and labels
images_all = [torch.randn(3, 4) for _ in range(8)]
labels_all = torch.tensor([0, 0, 0, 0, 1, 1, 1, 1], dtype=torch.long)

# Create indices_class structure (grouping sample indices by class)
indices_class = [
    [0, 1, 2, 3],  # class 0
    [4, 5, 6, 7]   # class 1
]

ARGS = {
    'images_all': images_all,
    'labels_all': labels_all,
    'indices_class': indices_class,
    'dataset': 'cifar10',
    'drop_criterion': 'random',
    'drop_ratio': 0.5,
    'keep_ratio': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
sample_indices_to_drop Ran
ran on a synthesized input
silicx/GoldFromOres-BiLP/drop_utils/drop.py
code served (permissive licence) · get_code("4a6b0c194587f5d1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned set. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import json
import os
import tempfile
import numpy as np

# Create a temporary directory for drop_utils/resource
temp_dir = tempfile.mkdtemp()
resource_dir = os.path.join(temp_dir, 'drop_utils', 'resource')
os.makedirs(resource_dir, exist_ok=True)

# Create a sample utility scores JSON file for the dataset
utility_scores = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0]
score_file_path = os.path.join(resource_dir, 'mnist_LossConverge.json')
with open(score_file_path, 'w') as f:
    json.dump(utility_scores, f)

# Create indices_class: a list of lists where each inner list contains sample indices for a class
# Class 0: indices 0-4, Class 1: indices 5-9
indices_class = [
    [0, 1, 2, 3, 4],
    [5, 6, 7, 8, 9]
]

ARGS = {
    'dataset': 'mnist',
    'drop_criterion': 'LossConverge_Small',
    'indices_class': indices_class,
    'keep_percentile_from': 0.2,
    'keep_percentile_to': 0.8,
    'CRITERION_ROOT': resource_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_dataset Not yet run
unverified (harvested, no recorded run)
silicx/GoldFromOres-BiLP/DatasetCondensation/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from VincenDen/IID, is pointer only (licence: NONE) · get_code("4ede4e38dc61074f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from VincenDen/IID, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

get_network Not yet run
unverified (harvested, no recorded run)
silicx/GoldFromOres-BiLP/DatasetCondensation/utils.py
code served (permissive licence) · get_code("a6dd063f2c93ac08")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 0 the same month, 4 later) and 0 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
VGG11BN 28 other papers
15 dated earlier, 2 later, 11 same year, order not determinable
2203.01531 (2022-03)
CAFE: Learning to Condense Dataset by Aligning Features
2204.07513 (2022-04)
Synthesizing Informative Training Samples with GAN
2210.16774 (2022-10)
Dataset Distillation via Factorization
and 25 more not named here
VGG11 27 other papers
15 dated earlier, 2 later, 10 same year, order not determinable
2203.01531 (2022-03)
CAFE: Learning to Condense Dataset by Aligning Features
2204.07513 (2022-04)
Synthesizing Informative Training Samples with GAN
2210.16774 (2022-10)
Dataset Distillation via Factorization
and 24 more not named here
VGG13 19 other papers
12 dated earlier, 1 later, 6 same year, order not determinable
2204.07513 (2022-04)
Synthesizing Informative Training Samples with GAN
2210.16774 (2022-10)
Dataset Distillation via Factorization
2211.04446 (2022-11)
Private Set Generation with Discriminative Information
and 16 more not named here
distance_wb 11 other papers
8 dated earlier, 1 later, 2 same year, order not determinable
2202.02916 (2022-02)
Dataset Condensation with Contrastive Signals
2204.07513 (2022-04)
Synthesizing Informative Training Samples with GAN
2211.04446 (2022-11)
Private Set Generation with Discriminative Information
and 8 more not named here
get_dataset 5 other papers
3 dated earlier, 2 same year, order not determinable
2204.07513 (2022-04)
Synthesizing Informative Training Samples with GAN
2305.18381 (2023-05)
Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation
2310.03295 (2023-10)
Can pre-trained models assist in dataset distillation?
and 2 more not named here
drop_samples 1 other paper
1 dated earlier
2305.18381 (2023-05)
Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation
get_network 1 other paper
1 dated earlier
2305.18381 (2023-05)
Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation
sample_indices_to_drop 1 other paper
1 dated earlier
2305.18381 (2023-05)
Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Data-efficient learning has garnered significant attention, especially given the current trend of large multi-modal models. Recently, dataset distillation has become an effective approach by synthesizing data samples that are essential for network training. However, it remains to be explored which samples are essential for the dataset distillation process itself. In this work, we study the data efficiency and selection for the dataset distillation task. By re-formulating the dynamics of distillation, we provide insight into the inherent redundancy in the real dataset, both theoretically and empirically. We propose to use the empirical loss value as a static data pruning criterion. To further compensate for the variation of the data value in training, we find the most contributing samples based on their causal effects on the distillation. The proposed selection strategy can efficiently exploit the training dataset, outperform the previous SOTA distillation algorithms, and consistently enhance the distillation algorithms, even on much larger-scale and more heterogeneous datasets, e.g., full ImageNet-1K and Kinetics-400. We believe this paradigm will open up new avenues in the dynamics of distillation and pave the way for efficient dataset distillation. Our code is available on https://github.com/silicx/GoldFromOres-BiLP.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("03030")
get_code_for_paper("03030")
have("03030")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/03030.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/03030.svg)](https://syntology.ai/paper/03030)

Connect an agent — have() is free.